Some special-purpose filters (such as the polarizers) did not have flats made until well after the installation of WFPC2. Some, such as the Woods filter (F160BW), still do not have an accurate flatfield. In these cases, the PEDIGREE and DESCRIP keywords for the flatfield file, reported at the end of the header of the calibrated image in HISTORY records, will indicate the existence of a problem at the time the data were taken. A pedigree of DUMMY indicates that the flatfield is identically 1, thus no real flatfield correction has been performed. A pedigree of GROUND indicates that the flatfield has been obtained purely from data taken during the pre-launch Thermal Vacuum test, and thus does not reflect properly the illumination function typical of the HST OTA; in this case, zonal errors of several percent can be encountered. Such cases are extremely rare after the summer of 1996, but are not uncommon for data taken in 1994 and early 1995, when the basic camera calibration was not yet completed.
A completely new generation of flatfields has been delivered between late 1995 (for the filters used in Hubble Deep Field observations) and 1996. These flatfields are more accurate than the previous ones, although few users will need to recalibrate as a result of this change. In the optical, the new flats differ from the old by 1% or less over the vast majority of the chip, with the differences growing in the outer 50 pixels of the chip to about 8% at all wavelengths. Longward of 850 nm, differences of up to 1.5% are seen across the main body of the chips, and shortward of 300 nm the differences between the old and new flatfields are less than 3% over most of the chip.
WFPC2 flatfields are defined so that a source of constant brightness produces the same count rate per pixel across the image. However, due to geometric distortion of the image by the optics, the area of WFPC2 pixels on the sky depends on location on the chip. The total variation across the chip is a few percent. Therefore, the photometry of point sources is slightly corrupted by the standard flattening procedure. This effect, and its correction, are discussed in "Geometric Distortion" on page 28-12.